Positron emission tomography (PET) is a functional imaging modality that is capable of imaging biochemical processes in humans or animals through the use of radioactive tracers. PET/CT with [18F]fluorodeoxyglucose (FDG) is increasingly being used for staging, restaging and treatment monitoring for cancer patients with different types of tumors. However, current FDG-PET provides a low sensitivity to detect micrometastases and small tumor infiltrated lymph nodes. The goal of this project is to improve the efficacy of PET imaging through the development of novel image reconstruction methods and data analysis tools. During the current funding period, we have developed a method for tuning reconstruction algorithm based on the noise characteristics in measured patient data. This patient-adaptive reconstruction algorithm has been validated using computer simulations and phantom experiments. In the next phase of the project, we will implement the patient-adaptive algorithm on clinical PET scanners and validate the method using patient data. We will further expand the capability of PET imaging by developing novel methods to utilize the anatomical information provided by PET/CT scanners and by exploring the potential of dynamic PET for cancer detection and staging. The four specific aims of the project are (1) To implement the patient-adaptive MAP reconstruction on clinical scanners and to validate the algorithm using breast cancer patients; (2) To develop a novel approach to PET image reconstruction using anatomical information; (3) To develop statistically efficient image reconstruction methods for dynamic PET; (4) To identify spatial-temporal features in dynamic PET image for detection and characterization of breast cancer and to evaluate the performance using breast cancer patient data. The first aim is an important step towards translating image reconstruction technology development into patient health care. Once validated using breast cancer patients, the method is readily applicable to imaging other types of tumors. The second to the fourth aims will greatly enhance the capability of PET by taking advantage of the recent advances in instrumentation (wide availability of PET/CT) and the dynamic nature of PET imaging. We expect the new methods to be developed will be able to extract clinically relevant features from dynamic PET for detecting small tumors and characterizing the malignancy of primary tumors. All the results will be validated using breast cancer patient data with histologically verified ground truth. The success of this research will have a significant and positive impact on the management of patients with breast cancer. PUBLIC HEALTH RELEVANCE: Positron emission tomography (PET) is a medical imaging technique that can detect cancer and monitor treatment response. This research aims to improve the efficacy of PET by developing novel image reconstruction and data processing tools that will enable early detection and characterization of breast cancer. The success of this research is of substantial benefit to the general population of breast cancer suffers.